.png){kind=icon}
.png){kind=icon}
Return
A Web-Based Platform for Korean Calligraphy Learners Using Deep Learning
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2025.3612578.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Calligraphy writing has a positive impact on multiple aspects of life. Learning to write in the calligraphic form helps in information acquisition, advancing language fluency, well-being development and unique personality disclosure. However, the trend of calligraphy writing is declining, and the younger generation tends to type with the keyboard instead of handwriting. This phenomenon has led to issues such as inappropriate handwriting. Therefore, a smart writing platform is required to provide writing practice to help the young generation with handwriting due to the inability to detect the correctness of handwriting. With the intense interest in deep learning, a writing platform is proposed which integrates deep learning technologies such as Convolutional Neural Network (CNN) and RetinaNet to resolve the issue of poor handwriting. The proposed platform is a beginner-friendly application that reduces the learning curve of Korean calligraphy writing, provides an interactive writing platform that allows users to practice, provides detection of handwriting correctness to train the users to get the correct writing and tracks the users’ learning progress. Deep learning model is employed for handwriting-detection module to accurately detect the correctness of a handwriting. Writing practice with varying levels of difficulty is delivered to the users. Users’ data and progress in writing practices are expected to be stored in the cloud service safely. The expected system outcome is a web-based writing platform that is able to display the correctness of handwriting input by the users with a mean accuracy, precision, recall and F1-score of 86.67%, 89.12%, 90.27% and 89.64%, respectively.
Keywords:
Convolutional neural network
deep learning
handwriting
Korean calligraphy
RetinaNet